On 08/06/11 08:00, Frank von Delft wrote:
> Surely using an absolute sigma cut-off is not a very sensible thing to
> do - given that sigma is a function of so many different things that
> have nothing to do with noise?
If you have val in electrons/A^3, then use (/ val (map-sigma
(imol-refinement-map)) as the argument to the function.
I will be renaming the "sigma" functions -> "rmsd" in 0.7.
>
> (Has anybody explored using local non-protein rmsd as sigma?)
I have not.
|